How can we avoid traffic jams? Design of on-demand traffic guidance systems II

• Main Author: Ramamoorthy, Gautham
• Other Authors: Justin Dauwels
• Format: Final Year Project
• Language: English
• Published: 2013
• Subjects: DRNTU::Engineering
• Online Access: http://hdl.handle.net/10356/54497
• Institution: Nanyang Technological University

With the increasing occurrence of traffic congestions, there is a growing need for traffic guidance systems that can accurately predict future traffic flows and guide users to take the optimal route. In this thesis, we primarily analyse a traffic data set obtained from LTA Singapore that contains speed values, to understand and exploit internal trends; which can then be applied to build practical transportation algorithms for the required traffic guidance system. To reduce the complexity of the analysis, popular term reduction techniques such as Principal Component Analysis (PCA) and Independent Component Analysis (ICA) are applied to the traffic data and their compression performances are compared. A new term reduction technique employing both PCA & ICA is postulated and its compression performance is measured. The report provides the theoretical background behind the different reduction techniques and presents step by step the method used for performing them through MATLAB.